The Tsinghua University Campus Map is a digital map service designed to provide users with an efficient way to search for Points of Interest (POI) within the campus. As part of this project, I utilized data mining techniques to enhance the search functionality and user experience.
Using the Generalized Sequential Pattern algorithm, I analyzed the dataset of search queries to discover frequent keyword patterns commonly used by users. By identifying these patterns, I was able to gain insights into user behavior, including common abbreviations and typos when searching for specific locations.
Furthermore, I explored association rules between keyword patterns and the search results clicked by users. These association rules played a crucial role in improving search suggestions and optimizing the ranking of search results. By leveraging the discovered patterns and rules, the search service could provide more accurate spell/query suggestions and deliver more relevant search results to users.
Throughout the project, my skills in Python, Numpy, data visualization using Matplotlib and Seaborn, sequential pattern discovery, association rules, and search engine optimization were instrumental in the successful implementation of the enhanced search functionality for the Tsinghua University Campus Map.
Highlights of this project include the discovery of frequent user input patterns, the establishment of association rules between patterns and search results, and the subsequent improvement of search suggestions and ranking. These advancements significantly enhanced the overall user experience and efficiency of the Tsinghua University Campus Map.